You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/server/namenode/NNThroughputBenchmark.java
+17-9Lines changed: 17 additions & 9 deletions
Original file line number
Diff line number
Diff line change
@@ -106,9 +106,9 @@
106
106
* By default the refresh is never called.</li>
107
107
* <li>-keepResults do not clean up the name-space after execution.</li>
108
108
* <li>-useExisting do not recreate the name-space, use existing data.</li>
109
-
* <li>-namenode will run the test against a namenode in another
110
-
* process or on another host. If you use this option, the namenode
111
-
* must have dfs.namenode.fs-limits.min-block-size set to 16.</li>
109
+
* <li>-namenode will run the test (except {@link ReplicationStats}) against a
110
+
* namenode in another process or on another host. If you use this option,
111
+
* the namenode must have dfs.namenode.fs-limits.min-block-size set to 16.</li>
112
112
* </ol>
113
113
*
114
114
* The benchmark first generates inputs for each thread so that the
@@ -126,8 +126,9 @@ public class NNThroughputBenchmark implements Tool {
0 commit comments